Archives for the month of: November, 2016

I love the blebby membrane on this cell.  Blebs (the bubbly structures) can be a result of apoptosis, otherwise known as programmed cell death, where the membrane blebs start budding off into space, filled with cell junk.  However in this case this case I think this type of cultured cells (CHO cells: Chinese Hamster Ovary)  just sort of have a bubbly membranes a lot; they are not apoptotic.

jh0005_40deg_4-5k_39mm_x

_mg_0171_x

The 3d reconstruction is made based on stereo pairs of images: two images at 5-10 degrees difference in tilt that could be viewed through a stereoscope, or computed into a 3d model.  Typically the reconstruction is done from two images near the top, with this cell the reconstruction was from two images on one side.  You can see the resulting distortion below.  But still looks like the cell when viewing from the angle it was reconstructed from.

_mg_0172_x

_mg_0155

For my independent study I am learning some lighting techniques, so I decided to apply them to my 3d printed cells in an attempt to make them look like an electron micrograph.  In scanning electron microscopy everything is grey and opaque, thus 3d printed cells have a good possibility to be rendered like an electron micrograph. I went with a gigantic softbox on one side, smaller softbox on the other side; to make the light wrap around.  Not entirely happy with the result, and definitely should have gone with a grey background instead of black.

thumbnail_img4-actual-cells

Original^

_mg_0169_x

 

3D reconstructed^

Electrons don’t quite behave like light- for one thing the detector has a positive voltage to draw electrons toward it, thus one can have a some electrons being sucked up from behind objects.  My professor suggested I combine multiple exposures of different lighting arrangements to try and get it to outline the forms like the top electron microscope image.  Below are the results- still not quite like electron microscopy.  _MG_9786 cropped.jpeg_MG_9785.JPG

_mg_0192_x^In the next iteration, I made a 3d printed mold of the cells, and cast them in clear silicone (it’s called ‘sorta-clear’, no lie).

I was interested in making an electron micrograph object be transparent, closer to a transmitted light image like below:x-polyarthra-vulgaris-male_11b-8

In the final iteration, not sure if I want to go there, but I’m interested in reproducing the look of what’s called Differential Interference Contrast (DIC) microscopy.  Things look like reliefs in DIC:

x-axiovert_a1

So I thought of doing a relief of cells, in yet another iterative reconstruction step.  The image below is a result of 3D reconstructing from the 2D transparent cell photo above, could be 3d printed.

untitled_x

I have the frame and bed assembled, this week I’ve been testing the electronics.  I decided to go with 24V electronics for the option of having more power to heat up my gigantic 200x300mm heated bed (this one: https://www.amazon.com/gp/product/B0158N171O/ref=oh_aui_search_detailpage?ie=UTF8&psc=1).  The bed is actually a bit confusing: basically it can be dual voltage (12 or 24V), and each voltage has a different resistance for the heating element- 1 ohm for 12V, 4 ohms for 24V.  If you do the math the power is about the same for each: 140W.  However, one can run 24V through the 12V terminal (see below), making the resistance 1 ohm- 1 ohm @24V= 575W! (25 amps, yikes).  A little overkill, but I can’t find a bed between 140 and 575W.  So this week I tested the bed using 24V in the 24V terminal, to see if it would reach ABS temps @140W.  After about 12 minutes it is stuck at 100 degrees.  Might be enough, but barely, so I decided to go with the 575W approach- I am now waiting on a relay and a separate power supply.

heatedbedlarge4

For my controller board, I am using a smoothieboard instead of RAMPs, mainly since I’ve heard good things and it works at 24V.
fullsizerender-2

It’s been a bit of an adventure figuring it out, the documentation is ok after you get to know it better, but at first I was lost.  But the customer service is great- they typically reply within an hour to my dumb questions.  The scary image below and the configuration code is actually starting to make sense, woohoo.

untitled

img_7967For my ‘Machines that Make’ class I have started a d-bot 3d printer, as found here: http://www.thingiverse.com/thing:1001065.  I was sold on it because of the great documentation and sturdy frame.  However I keep making mistakes on the details like ordering the wrong gauge wire (and cutting it, so I can’t return it), so it’s getting a bit expensive, oh well.  img_7971